Cloud Storage for Snapshots
Kubeshark can upload and download Traffic Snapshots to cloud object storage, enabling cross-cluster sharing, backup/restore, and long-term retention.
Supported providers: Amazon S3 (s3) and Azure Blob Storage (azblob).
Helm Values
tap:
snapshots:
cloud:
provider: "" # "s3" or "azblob" (empty = disabled)
prefix: "" # key prefix in the bucket/container (e.g. "snapshots/")
configMaps: [] # names of pre-existing ConfigMaps with cloud config env vars
secrets: [] # names of pre-existing Secrets with cloud credentials
s3:
bucket: ""
region: ""
accessKey: ""
secretKey: ""
roleArn: ""
externalId: ""
azblob:
storageAccount: ""
container: ""
storageKey: ""
providerselects which cloud backend to use. Leave empty to disable cloud storage.configMapsandsecretsare lists of names of existing ConfigMap/Secret resources. They are mounted asenvFromon the hub pod, injecting all their keys as environment variables.
Inline Values (Alternative to External ConfigMaps/Secrets)
Instead of creating ConfigMap and Secret resources manually, you can set cloud storage configuration directly in values.yaml or via --set flags. The Helm chart will automatically create the necessary ConfigMap and Secret resources.
Both approaches can be used together — inline values are additive to external configMaps/secrets references.
Inline Helm value → Environment variable mapping:
| Helm Value | Auto-Created Resource | Environment Variable |
|---|---|---|
tap.snapshots.cloud.prefix | ConfigMap | SNAPSHOT_CLOUD_PREFIX |
tap.snapshots.cloud.s3.bucket | ConfigMap | SNAPSHOT_AWS_BUCKET |
tap.snapshots.cloud.s3.region | ConfigMap | SNAPSHOT_AWS_REGION |
tap.snapshots.cloud.s3.roleArn | ConfigMap | SNAPSHOT_AWS_ROLE_ARN |
tap.snapshots.cloud.s3.externalId | ConfigMap | SNAPSHOT_AWS_EXTERNAL_ID |
tap.snapshots.cloud.s3.accessKey | Secret | SNAPSHOT_AWS_ACCESS_KEY |
tap.snapshots.cloud.s3.secretKey | Secret | SNAPSHOT_AWS_SECRET_KEY |
tap.snapshots.cloud.azblob.storageAccount | ConfigMap | SNAPSHOT_AZBLOB_STORAGE_ACCOUNT |
tap.snapshots.cloud.azblob.container | ConfigMap | SNAPSHOT_AZBLOB_CONTAINER |
tap.snapshots.cloud.azblob.storageKey | Secret | SNAPSHOT_AZBLOB_STORAGE_KEY |
Amazon S3
Environment Variables
| Variable | Required | Description |
|---|---|---|
SNAPSHOT_AWS_BUCKET | Yes | S3 bucket name |
SNAPSHOT_AWS_REGION | No | AWS region (uses SDK default if empty) |
SNAPSHOT_AWS_ACCESS_KEY | No | Static access key ID (empty = use default credential chain) |
SNAPSHOT_AWS_SECRET_KEY | No | Static secret access key |
SNAPSHOT_AWS_ROLE_ARN | No | IAM role ARN to assume via STS (for cross-account access) |
SNAPSHOT_AWS_EXTERNAL_ID | No | External ID for the STS AssumeRole call |
SNAPSHOT_CLOUD_PREFIX | No | Key prefix in the bucket (e.g. snapshots/) |
Authentication Methods
Credentials are resolved in this order:
- Static credentials — If
SNAPSHOT_AWS_ACCESS_KEYis set, static credentials are used directly. - STS AssumeRole — If
SNAPSHOT_AWS_ROLE_ARNis also set, the static (or default) credentials are used to assume the given IAM role. This is useful for cross-account S3 access. - AWS default credential chain — When no static credentials are provided, the SDK default chain is used:
- IRSA (EKS service account token) — recommended for production on EKS
- EC2 instance profile
- Standard AWS environment variables (
AWS_ACCESS_KEY_ID,AWS_SECRET_ACCESS_KEY, etc.) - Shared credentials file (
~/.aws/credentials)
The provider validates bucket access on startup via HeadBucket. If the bucket is inaccessible, the hub will fail to start.
Examples Using Inline Values
Basic S3 (IRSA / default credentials)
When running on EKS with IRSA or another default credential method, only bucket and region are needed:
tap:
snapshots:
cloud:
provider: "s3"
s3:
bucket: my-kubeshark-snapshots
region: us-east-1
The hub pod’s service account must be annotated for IRSA with an IAM role that has S3 access to the bucket.
S3 with Static Credentials
tap:
snapshots:
cloud:
provider: "s3"
s3:
bucket: my-kubeshark-snapshots
region: us-east-1
accessKey: AKIA...
secretKey: wJal...
Or equivalently via --set:
helm install kubeshark kubeshark/kubeshark \
--set tap.snapshots.cloud.provider=s3 \
--set tap.snapshots.cloud.s3.bucket=my-kubeshark-snapshots \
--set tap.snapshots.cloud.s3.region=us-east-1 \
--set tap.snapshots.cloud.s3.accessKey=AKIA... \
--set tap.snapshots.cloud.s3.secretKey=wJal...
S3 Cross-Account Access via AssumeRole
tap:
snapshots:
cloud:
provider: "s3"
s3:
bucket: other-account-bucket
region: eu-west-1
roleArn: arn:aws:iam::123456789012:role/KubesharkCrossAccountRole
externalId: my-external-id # optional, if required by the trust policy
The hub will first authenticate using its own credentials (IRSA, static, or default chain), then assume the specified role to access the bucket.
S3 with Prefix
Use prefix to namespace snapshot keys within a bucket:
tap:
snapshots:
cloud:
provider: "s3"
prefix: "production/snapshots"
s3:
bucket: my-kubeshark-snapshots
region: us-east-1
Examples Using External ConfigMaps/Secrets
If you prefer to manage credentials as separate Kubernetes resources (e.g. managed by an external secrets operator), use the configMaps and secrets fields instead of inline values.
IRSA (recommended for EKS)
Create a ConfigMap with bucket configuration:
apiVersion: v1
kind: ConfigMap
metadata:
name: kubeshark-s3-config
data:
SNAPSHOT_AWS_BUCKET: my-kubeshark-snapshots
SNAPSHOT_AWS_REGION: us-east-1
Set Helm values:
tap:
snapshots:
cloud:
provider: "s3"
configMaps:
- kubeshark-s3-config
The hub pod’s service account must be annotated for IRSA with an IAM role that has S3 access to the bucket.
Static Credentials
Create a Secret with credentials:
apiVersion: v1
kind: Secret
metadata:
name: kubeshark-s3-creds
type: Opaque
stringData:
SNAPSHOT_AWS_ACCESS_KEY: AKIA...
SNAPSHOT_AWS_SECRET_KEY: wJal...
Create a ConfigMap with bucket configuration:
apiVersion: v1
kind: ConfigMap
metadata:
name: kubeshark-s3-config
data:
SNAPSHOT_AWS_BUCKET: my-kubeshark-snapshots
SNAPSHOT_AWS_REGION: us-east-1
Set Helm values:
tap:
snapshots:
cloud:
provider: "s3"
configMaps:
- kubeshark-s3-config
secrets:
- kubeshark-s3-creds
Cross-Account Access via AssumeRole
Add the role ARN to your ConfigMap:
apiVersion: v1
kind: ConfigMap
metadata:
name: kubeshark-s3-config
data:
SNAPSHOT_AWS_BUCKET: other-account-bucket
SNAPSHOT_AWS_REGION: eu-west-1
SNAPSHOT_AWS_ROLE_ARN: arn:aws:iam::123456789012:role/KubesharkCrossAccountRole
SNAPSHOT_AWS_EXTERNAL_ID: my-external-id # optional, if required by the trust policy
The hub will first authenticate using its own credentials (IRSA, static, or default chain), then assume the specified role to access the bucket.
Azure Blob Storage
Environment Variables
| Variable | Required | Description |
|---|---|---|
SNAPSHOT_AZBLOB_STORAGE_ACCOUNT | Yes | Azure storage account name |
SNAPSHOT_AZBLOB_CONTAINER | Yes | Blob container name |
SNAPSHOT_AZBLOB_STORAGE_KEY | No | Storage account access key (empty = use DefaultAzureCredential) |
SNAPSHOT_CLOUD_PREFIX | No | Key prefix in the container (e.g. snapshots/) |
Authentication Methods
Credentials are resolved in this order:
- Shared Key — If
SNAPSHOT_AZBLOB_STORAGE_KEYis set, the storage account key is used directly. - DefaultAzureCredential — When no storage key is provided, the Azure SDK default credential chain is used:
- Workload Identity (AKS pod identity) — recommended for production on AKS
- Managed Identity (system or user-assigned)
- Azure CLI credentials
- Environment variables (
AZURE_CLIENT_ID,AZURE_TENANT_ID,AZURE_CLIENT_SECRET)
The provider validates container access on startup via GetProperties. If the container is inaccessible, the hub will fail to start.
Examples Using Inline Values
Workload Identity (recommended for AKS)
When using AKS Workload Identity or another default credential method, only the storage account and container are needed:
tap:
snapshots:
cloud:
provider: "azblob"
azblob:
storageAccount: mykubesharksa
container: snapshots
The hub pod’s service account must be configured for AKS Workload Identity with a managed identity that has the Storage Blob Data Contributor role on the container.
Storage Account Key
tap:
snapshots:
cloud:
provider: "azblob"
azblob:
storageAccount: mykubesharksa
container: snapshots
storageKey: "base64-encoded-storage-key..."
Examples Using External ConfigMaps/Secrets
Workload Identity (recommended for AKS)
Create a ConfigMap with storage configuration:
apiVersion: v1
kind: ConfigMap
metadata:
name: kubeshark-azblob-config
data:
SNAPSHOT_AZBLOB_STORAGE_ACCOUNT: mykubesharksa
SNAPSHOT_AZBLOB_CONTAINER: snapshots
Set Helm values:
tap:
snapshots:
cloud:
provider: "azblob"
configMaps:
- kubeshark-azblob-config
The hub pod’s service account must be configured for AKS Workload Identity with a managed identity that has the Storage Blob Data Contributor role on the container.
Storage Account Key
Create a Secret with the storage key:
apiVersion: v1
kind: Secret
metadata:
name: kubeshark-azblob-creds
type: Opaque
stringData:
SNAPSHOT_AZBLOB_STORAGE_KEY: "base64-encoded-storage-key..."
Create a ConfigMap with storage configuration:
apiVersion: v1
kind: ConfigMap
metadata:
name: kubeshark-azblob-config
data:
SNAPSHOT_AZBLOB_STORAGE_ACCOUNT: mykubesharksa
SNAPSHOT_AZBLOB_CONTAINER: snapshots
Set Helm values:
tap:
snapshots:
cloud:
provider: "azblob"
configMaps:
- kubeshark-azblob-config
secrets:
- kubeshark-azblob-creds
Related Documentation
- Traffic Snapshots — Overview of snapshot capabilities
- Snapshots Dashboard — Creating and managing snapshots
- Dashboard Cloud Storage — Connection status and location filtering
- Helm Configuration Reference — All snapshot Helm values
- Raw Capture Configuration — Local storage and capture settings
- MCP Cloud Storage Tools — Upload/download snapshots via AI assistants